[00:00:00] Speaker 01: The next case for argument is 23-1428, Egenera v. Sisko. [00:00:08] Speaker 01: Mr. Brunelli, whenever you're ready. [00:00:10] Speaker 03: Good morning, and may it please the Court. [00:00:12] Speaker 03: My name is Robert Brunelli, and I represent Egenera. [00:00:15] Speaker 03: Egenera asserted independent claims 1, 3, 5, and 7 of the 430 patent against Cisco's unified computing system, sometimes called UCS. [00:00:28] Speaker 03: District Court granted summary judgment of non-infringement of claims 1 and 5, finding that eGenera failed to generate evidence showing that the CPUs, which are included as part of each Cisco UCS, participated in emulation of Ethernet functionality that is provided by the UCSs. [00:00:56] Speaker 03: Claims 3 and 7 were tried to the jury. [00:00:58] Speaker 03: The jury found those claims not infringed and not invalid. [00:01:03] Speaker 03: He generally appealed, seeking reversal of the summary judgment order in a remand. [00:01:07] Speaker 04: Can you talk about the summary judgment piece? [00:01:12] Speaker 04: I'm not sure which side here is causing me greater concern about that. [00:01:20] Speaker 04: The reason I say that is that what [00:01:24] Speaker 04: least to me is strikingly absent is a meaningful explanation from experts from wherever of what under the hood concretely it means to emulate Ethernet functionality. [00:01:45] Speaker 04: And without that, when I hear this debate, they say and the judge said, well, just using it [00:01:54] Speaker 04: using Ethernet functionality is not emulating it. [00:01:58] Speaker 04: And you say, well, yeah, it is. [00:02:01] Speaker 04: And I say, OK, how do I resolve that? [00:02:04] Speaker 04: And I'm not informed about what emulating Ethernet functionality is. [00:02:11] Speaker 04: I need some light shed on this. [00:02:18] Speaker 02: And ideally, if it's in the record, that'd be even better. [00:02:25] Speaker 03: I would direct the court. [00:02:27] Speaker 03: So in context, so what does emulate ethernet functionality mean? [00:02:34] Speaker 03: Like in Toto, it means all sorts of things. [00:02:37] Speaker 03: But in the context of this summary judgment motion, the dispute seems to be at one level. [00:02:45] Speaker 03: Does being able to communicate from the CPU using the MAC address to other components of the CS [00:02:55] Speaker 03: of the UCS is that being able to communicate peace part of emulation of Ethernet functionality. [00:03:10] Speaker 04: claim element that I think is one of the issues on the JMAW. [00:03:13] Speaker 04: This is not one where the claim element says the logic has to be establishing the ability to use ethernet to emulate functionality. [00:03:28] Speaker 04: It's just emulating. [00:03:31] Speaker 04: Unlike this, whatever the other, about the network topology or something. [00:03:35] Speaker 04: Establishing the network topology. [00:03:40] Speaker 04: But I want to stick with the emulate part. [00:03:43] Speaker 04: Yes. [00:03:44] Speaker 03: But what are we emulating? [00:03:45] Speaker 03: We're emulating, if we look at claim five, for instance. [00:03:50] Speaker 04: Let me just say, I mean, it may well be that your answer to this is that without an explanation, you can't do summary judgment on this. [00:03:59] Speaker 03: That is part of my argument, yes. [00:04:05] Speaker 02: So wait, does that mean you're conceding there is no explanation in the record? [00:04:10] Speaker 03: There's no explanation in the court's order on the summary judgment. [00:04:15] Speaker 01: Did anyone ask for claim construction? [00:04:19] Speaker 03: Interestingly, aspects of this particular term were submitted to claim construction. [00:04:26] Speaker 03: Just the processor part. [00:04:27] Speaker 03: Correct. [00:04:29] Speaker 04: No, there was claim construction on whether... The later clause, the later words, the internal communication network. [00:04:37] Speaker 04: That's what there's a page and a half in the claim construction opinion about, but not about what it means to emulate Ethernet functionality. [00:04:45] Speaker 01: That is absolutely true, Your Honor. [00:04:46] Speaker 01: So my question remains, did anyone request claim construction? [00:04:51] Speaker 02: and you haven't appealed the lack of a claim construction, correct? [00:04:55] Speaker 03: Correct. [00:04:55] Speaker 02: So I don't see how this provides a basis for reversing summary judgment. [00:04:59] Speaker 02: Do you have an answer? [00:05:00] Speaker 03: Yes. [00:05:01] Speaker 03: So if there's no claim construction, we're looking at plain and ordinary meaning. [00:05:05] Speaker 03: And if we're looking at plain and ordinary meaning, we look at the specification. [00:05:09] Speaker 03: The best evidence of what the plain and ordinary meaning of terms in the claims are is the specification to the patent. [00:05:16] Speaker 03: If we look at appendix 103, which is [00:05:21] Speaker 03: columns three and four of the 430 patents. [00:05:28] Speaker 03: We see at lines 55 to 60 a discussion of, quote, each pane through software commands is configured to have a corresponding subset of processors that may communicate via the virtual local area network, that is, [00:05:49] Speaker 03: emulated over the PTP mesh. [00:05:54] Speaker 03: And it continues. [00:05:55] Speaker 02: So, what is the plain letter meaning of emulating the Ethernet? [00:06:02] Speaker 03: Well, the evidence in the record is that the MAC addresses are programmed into the CPUs and that they are then used to communicate as part of the overall PAM. [00:06:18] Speaker 03: And so, [00:06:19] Speaker 04: Just to be, I know that there was, at least in my mind, there was some confusion about this. [00:06:25] Speaker 04: You're not any longer disputing, if you ever were, that the CPU does not itself get assigned a MAC address. [00:06:34] Speaker 04: But you're saying that the CPU has MAC addresses of other components within the system that are assigned them. [00:06:42] Speaker 04: And the CPU, when it sends messages out, will [00:06:47] Speaker 04: use those MAC addresses in sending those messages out? [00:06:52] Speaker 04: Is that a correct understanding? [00:06:54] Speaker 03: It is correct. [00:06:55] Speaker 03: One small modification. [00:06:59] Speaker 03: The MAC address is the MAC address of the universal computing system. [00:07:06] Speaker 03: So you got the board, you got four or six CPUs on it. [00:07:11] Speaker 03: Each of those CPUs gets the same MAC address because it's all one MAC address assigned to that. [00:07:17] Speaker 04: OK, which is not different from the larger outfit of which it's a part that has the MAC address. [00:07:26] Speaker 03: Correct. [00:07:26] Speaker 03: OK. [00:07:41] Speaker 04: Can I just ask you, I guess, about the JMA part of this? [00:07:45] Speaker 04: In a way, at least, I've been thinking about this as sort of presenting something like the same issue. [00:07:53] Speaker 04: That is, for each of the three claim terms that you're focusing on here, the claim construction takes one only so far. [00:08:03] Speaker 04: And then in that circumstance, it is the rule, which we've articulated lots of times, that [00:08:11] Speaker 04: when there's no longer a dispute about claim construction than a jury, then the fact findings under it are permissible as long as they are [00:08:22] Speaker 04: one reasonable application of either the unconstrued term or the construction that it is not further construed. [00:08:32] Speaker 04: And it seems to me that gives you a problem for each of the three JML issues. [00:08:38] Speaker 04: It may actually be helpful to you on the summary judgment point, because without a construction of emulating ethernet functionality, anything that's reasonable [00:08:50] Speaker 04: would be a jury issue. [00:08:53] Speaker 04: But now on the Jamal part, why isn't each of the conclusions or findings that we attribute to the jury about the address and about the modifying, and I don't remember what the third one was, within the range of the reasonable as an application of the claim limitation that's at issue? [00:09:23] Speaker 04: Sorry, there was a lot in that question. [00:09:28] Speaker 03: No, that was good, though. [00:09:31] Speaker 03: So what I want to say is that it's almost jumping over your question and going to the request for a new trial, saying the jury got confused. [00:09:42] Speaker 04: I want to concentrate on the Jamal evidence. [00:09:46] Speaker 03: So I think you raise a good point. [00:09:50] Speaker 03: Those arguments really come down to the claim construction issues. [00:09:53] Speaker 03: What do those terms mean? [00:09:55] Speaker 03: And because it's a plain and ordinary meaning, can we accept what the jury decided? [00:10:03] Speaker 03: Yeah. [00:10:03] Speaker 03: I think you probably can in that context. [00:10:09] Speaker 03: But we have a general verdict form here, not a specific. [00:10:14] Speaker 03: So we don't know which of those three, [00:10:17] Speaker 03: All we know is mainly one of those three applied. [00:10:20] Speaker 02: Unless they got confused by... You have to show us that under the plain nori meaning, the jury could not reasonably, having taken all the evidence against you, could not reasonably reach any one of those three conclusions, correct? [00:10:33] Speaker 02: Correct. [00:10:34] Speaker 03: And that is in steep hill. [00:10:35] Speaker 02: Yeah. [00:10:35] Speaker 02: I don't think... Thank you for conceding that. [00:10:37] Speaker 02: Just one quick question on the new trial motion. [00:10:41] Speaker 02: I understand why you are unhappy with what the judge said. [00:10:46] Speaker 02: uh... in one sentence of his not even preliminary instructions but just what he said basically to welcome the jury and assure them what a patent is and he you know he colloquially said you know patent infringement is like copying something to that effect you now want to turn that into you say that set the tone for the entire trial and was foundational to the case those like handful of words [00:11:14] Speaker 02: before the jury even started, before preliminary instructions, before evidence, before final instructions. [00:11:19] Speaker 02: And yet you wait till right near the end of trial, 10 days later, to even ask for any cure. [00:11:26] Speaker 02: How could that possibly be an abuse of discretion and so prejudicial that we should reverse? [00:11:33] Speaker 03: So on the new trial issue, it's a totality of the circumstances standard, as I understand it, essentially. [00:11:40] Speaker 03: And so we have to look at everything. [00:11:42] Speaker 03: That was the first domino in the series of copying dominoes that were included in the overall case. [00:11:50] Speaker 03: We had that instruction, followed by an instruction that the court wouldn't give about if you've got your own patent, you can still infringe, which is a bad amount of content. [00:12:05] Speaker 02: But you properly instructed on what infringement is at length, correct? [00:12:09] Speaker 03: Correct. [00:12:11] Speaker 03: We didn't raise any issues with regard to the instructions he did give, correct? [00:12:16] Speaker 03: We've got the lay witnesses providing in the context of we didn't copy, we don't infringe. [00:12:23] Speaker 03: And then we've got a closing that uses the word copying over 30 times, more than infringement. [00:12:28] Speaker 01: Wasn't copying something that you put a lot in the record too throughout this trial? [00:12:33] Speaker 01: I mean, they weren't using this word copying. [00:12:35] Speaker 01: I thought it was central part of your case. [00:12:39] Speaker 02: You had a welfare infringement claim, right? [00:12:42] Speaker 02: He generally didn't. [00:12:43] Speaker 03: He generally absolutely didn't. [00:12:45] Speaker 02: Not surprising there was talking evidence. [00:12:47] Speaker 01: Correct. [00:12:50] Speaker 01: All right. [00:12:50] Speaker 01: Why don't we reserve the remainder of your photo and hear from the other side. [00:13:03] Speaker 01: Good morning. [00:13:03] Speaker 00: Good morning. [00:13:03] Speaker 00: May it please the court? [00:13:04] Speaker 00: Elizabeth Moulton for Cisco. [00:13:07] Speaker 00: Cisco's UCS system operates in fundamentally different ways than eGenera's claim system. [00:13:12] Speaker 00: And the main difference relevant here is that Cisco's system groups servers into virtual networks, whereas eGenera's claim system is about grouping processors into different networks. [00:13:23] Speaker 00: Because eGenera wanted to group processors together, its claims have really specific limitations about what needs to happen on the processors themselves to create their network. [00:13:34] Speaker 00: In contrast, Cisco's innovation was about how to group servers. [00:13:39] Speaker 00: So the logic and computer software for creating the server networks is found on our customized network interface cards that sit on each server and mediate between the processors and the fabric interconnect or the control node. [00:13:53] Speaker 00: That basic architectural difference meant that eGenera couldn't show infringement either at summary judgment or at trial. [00:14:01] Speaker 00: because Cisco doesn't put the network emulation? [00:14:03] Speaker 04: Yeah, talk about the emulation, because in particular, I guess the way that I've been thinking about it led me to be struck by your use several times of expressions, just in your introduction here, about creating the network. [00:14:21] Speaker 04: which I understand is a fair description of the claim limitation about establishing the network topology, it doesn't feel to me like it's a fair description of the emulate term. [00:14:34] Speaker 00: Sure. [00:14:34] Speaker 00: So just to be clear, we're talking about both claims one and five. [00:14:38] Speaker 00: Yes. [00:14:39] Speaker 00: And claim one specifically has the logic to emulate. [00:14:41] Speaker 04: Neither side has argued any difference between having the logic to emulate and emulating. [00:14:50] Speaker 04: That's right. [00:14:51] Speaker 00: And the district court treated them the same at appendix 49. [00:14:55] Speaker 00: The district court says the claims require the CPUs include some logic to emulate Ethernet functionality or to emulate Ethernet functionality in some respect. [00:15:07] Speaker 00: So both claims 1 and claim 5 require that the processors be doing something to create or emulate the Ethernet functionality. [00:15:15] Speaker 00: Not create. [00:15:16] Speaker 00: to actually do the emulation. [00:15:18] Speaker 04: Do the emulation. [00:15:19] Speaker 04: So it comes down to what it means to emulate the functionality. [00:15:24] Speaker 04: And that's where I, early on, start looking for something to do some expert report, something that, or some claim construction analysis where it is explained and I come up empty. [00:15:40] Speaker 00: Sure. [00:15:41] Speaker 00: The plain meaning in light of the specification of emulating Ethernet. [00:15:47] Speaker 04: The plain is not going to do it. [00:15:49] Speaker 00: OK, fair enough. [00:15:51] Speaker 00: What the district court was confronted with at summary judgment is the evidence that EGENERA put forth, which is that 4127 is the excerpt of their expert report. [00:16:01] Speaker 00: That expert report says that there is logic to emulate the VNICs and related interfaces. [00:16:09] Speaker 00: So the VNIC is the network interface card. [00:16:11] Speaker 00: emulate Ethernet functionality. [00:16:13] Speaker 00: That's appendix 4127 paragraph 211. [00:16:17] Speaker 00: So that's the evidence that this report has. [00:16:19] Speaker 00: And Egenera says, OK, the VNICs emulate the processors used. [00:16:24] Speaker 04: And the distro... No, it didn't say that. [00:16:27] Speaker 04: Sorry? [00:16:28] Speaker 04: It didn't say the processors don't emulate. [00:16:30] Speaker 04: It says the processors do emulate. [00:16:33] Speaker 04: And how do they emulate? [00:16:34] Speaker 04: By communicating over and using. [00:16:37] Speaker 00: So the Egena's evidence was that the processors, sorry, their evidence is that the VNICs emulate. [00:16:44] Speaker 00: Their argument in their summary judgment brief at 52224 is that the processors use the emulated functionality to communicate over the emulated network. [00:16:57] Speaker 04: And that that constitutes emulating. [00:17:01] Speaker 00: that the use in communication constitutes emulating. [00:17:04] Speaker 00: That's their attorney argument. [00:17:05] Speaker 00: When the district court is confronted with that, he says, your evidence is the VNICs do the emulation. [00:17:12] Speaker 00: Your argument is that the CPUs use the emulated functionality created on the VNICs. [00:17:17] Speaker 00: That doesn't meet the claim language. [00:17:19] Speaker 00: There wasn't any factual dispute to resolve or any claim construction dispute. [00:17:23] Speaker 00: It was simply the district court saying, [00:17:25] Speaker 00: the evidence you put forth doesn't match the claim limitation, so there's no infringement, and that's a perfectly appropriate analysis at summary judgment. [00:17:33] Speaker 00: I can go through what is actually happening on the CPUs with the vNICS MAC address, if that would be helpful, but I think the basic point is just the evidence put forward didn't match the claim limitation, and because there wasn't any claim construction that would expand emulation to just include use of the functionality created by the NICS, there was no infringement. [00:17:56] Speaker 02: What if we were to now think that is a claim construction dispute, although I recognize no one asked the district court and no one is seemingly asking us, but what if we're left with that feeling nonetheless? [00:18:09] Speaker 00: So I don't think the court should engage in sort of suesponte claim construction or tell the district court it needed to do claim construction. [00:18:17] Speaker 00: EGENERA as the patentee should have raised that if they thought there was a dispute and that they thought use should encompass emulation. [00:18:24] Speaker 00: or emulation should encompass mere use? [00:18:27] Speaker 04: But I guess I think I tried to articulate this a little bit when your friend on the other side was speaking. [00:18:38] Speaker 04: When there is a term and its application is not yet clear from exactly what the claim construction has been or from the language if it hasn't been construed, same thing. [00:18:54] Speaker 04: then the rule is, we've said this a number of times, the LSI opinion last year, and a number of times, and it has to be for rule, right? [00:19:04] Speaker 04: If there's no further claim construction, then it's up to the jury to decide within the range of any reasonable application of this term. [00:19:21] Speaker 04: So I don't see how you can win. [00:19:24] Speaker 04: on that, if we think there's uncertainty about whether communicating over and using is emulation and something more, we don't have a reason yet to see why that position is just not reasonable under the language of emulate functionality. [00:19:45] Speaker 04: then you can't give summary judgment. [00:19:47] Speaker 04: You either need to go back and construe it and maybe then claim summary judgment is possible, but not without the construction. [00:19:57] Speaker 04: That's where I guess I'm stuck. [00:19:59] Speaker 00: Understood. [00:20:00] Speaker 00: So a couple of things. [00:20:04] Speaker 00: Originally leading up to summary judgment, the dispute over emulated Ethernet functionality [00:20:10] Speaker 00: was that Cisco's system actually uses Ethernet functionality. [00:20:14] Speaker 00: It doesn't emulate. [00:20:16] Speaker 00: So there was actually a claim construction dispute about whether directly using Ethernet constituted emulation. [00:20:22] Speaker 04: That's because in the internal communication network, your system actually uses Ethernet. [00:20:27] Speaker 00: And their system used the GigaNet network. [00:20:30] Speaker 00: So that really was the focus of discovery and some of the initial fights about what this claim term meant. [00:20:37] Speaker 00: When we got to summary judgment, [00:20:39] Speaker 00: At that point, Egenera's evidence was just that the emulation is happening on the NICs. [00:20:45] Speaker 00: And so we said, OK, that doesn't meet the claim limitation. [00:20:48] Speaker 00: And Egenera came back with, I'm not going to prove to you emulation happens on the CPU. [00:20:53] Speaker 00: I'm going to prove to you that use of this network is on the CPU. [00:20:57] Speaker 00: But by the time the CPUs are able to communicate over the network, the emulation has already happened. [00:21:04] Speaker 00: The whole point of Cisco's system is that [00:21:06] Speaker 00: this complicated emulation software and the logic is put on the NICs so that we don't have to put it on the processors. [00:21:14] Speaker 04: This might be a good place at least. [00:21:16] Speaker 04: I want to ask you one question and then I want to get back to something you offered to do. [00:21:21] Speaker 04: First of all, you have a paragraph I think in your brief that says, [00:21:25] Speaker 04: Well, because two of the three limitations that were the subject of their Jamal motion apply equally to claims one and five, then harmless error on the emulate, and I don't [00:21:44] Speaker 04: My problem with that is we don't know that the jury adopted either of the two that would do that for you. [00:21:51] Speaker 04: It might have ruled for you at the trial just on the third one, which doesn't do that for you. [00:21:56] Speaker 00: That's a problem. [00:21:57] Speaker 04: That's correct. [00:21:57] Speaker 04: Okay. [00:21:58] Speaker 04: So now I'm going to take you up on your offer to explain what under the hood is going on with the CPUs and the MAC addresses. [00:22:06] Speaker 04: Sure. [00:22:07] Speaker 00: So in the UCS system, the Fabric Interconnect is sort of the central brain, and that's the accused control node. [00:22:15] Speaker 00: You operate the Fabric Interconnect through a software program called UCS Manager, and you tell it, I want to have three servers in my virtual network. [00:22:24] Speaker 00: And it sends to three servers, OK, you're in the virtual network. [00:22:28] Speaker 00: It tells the NICs on those servers, this is your new MAC address. [00:22:34] Speaker 00: At that point, the network is established. [00:22:36] Speaker 00: created the server network that you want to have, and the NICs have their address. [00:22:44] Speaker 00: When the CPU, which sits behind the NIC on the server, is ready to send a message, it asks the NIC, what is your MAC address? [00:22:53] Speaker 00: I want to send a message, what is the from address I should use? [00:22:57] Speaker 00: And so the NIC says, OK, this is the MAC address that you're going to say your message is from. [00:23:01] Speaker 00: The message goes to the NIC. [00:23:03] Speaker 00: The NIC does all the complicated emulation programming, it adds tags, it reformats the message, and then it goes up to the Fabric Interconnect to get sent wherever it needs to get sent. [00:23:13] Speaker 04: Does it ever ask for the MAC address of the two, as opposed to the from? [00:23:20] Speaker 00: I think the NIC would supply that MAC address. [00:23:23] Speaker 00: The CPU would say where it wants the message to go. [00:23:26] Speaker 04: In what terms? [00:23:30] Speaker 00: In what terms. [00:23:31] Speaker 04: It's not going to give the non-MAC address location. [00:23:38] Speaker 00: Right. [00:23:39] Speaker 00: It depends if it's staying in the internal network or if it's going to an external storage or communication network. [00:23:46] Speaker 00: Both will happen, right? [00:23:47] Speaker 00: Yeah. [00:23:48] Speaker 00: Right. [00:23:48] Speaker 00: So that's why the DISH reports analogy. [00:23:50] Speaker 04: So whether it's to or from, why isn't the CPU [00:24:00] Speaker 04: logic instruction that asks for how to fill in, either the from or the to line or both, logic for emulating Ethernet functionality. [00:24:13] Speaker 04: Or I should say, why under what we currently have in the record, why isn't it reasonable for that characterization to be made? [00:24:27] Speaker 00: So what the CPU is [00:24:30] Speaker 00: isn't doing any of the networking itself. [00:24:33] Speaker 00: All of the networking is happening on the NIC. [00:24:36] Speaker 04: I don't know what the word networking means. [00:24:38] Speaker 00: Well, the ethernet networking, the ethernet emulation. [00:24:42] Speaker 04: That's just the conclusion. [00:24:45] Speaker 00: OK. [00:24:45] Speaker 00: So when you create the message on the CPU and it goes up to the NIC, the NIC has the programming and logic to actually emulate ethernet functionality. [00:24:59] Speaker 00: That is where everything gets packaged, put in its final package to go through Cisco's UCS system. [00:25:08] Speaker 00: What's happening on the processors is just the really basic functionality about computer, what computers always do, sending messages, storing things, things like that. [00:25:21] Speaker 00: And that's why the district court's phone analogy was appropriate here. [00:25:25] Speaker 00: So if you're creating a telephone network, [00:25:28] Speaker 00: You've got the telephones and telephone numbers that are set up. [00:25:32] Speaker 00: The telephone is like the NIC. [00:25:34] Speaker 00: It's assigned a phone number, and you can pick up the phone and use it and communicate over the phone, but the person that's using the phone sending messages isn't emulating any functionality of the telephone network. [00:25:46] Speaker 00: And that's effectively what the CPUs are doing here, is they're taking all the functionality created by the NIC and the control node and using that to communicate, but they themselves are not [00:25:57] Speaker 00: They don't have any of the logic or functionality that would allow them to communicate over an emulated Ethernet network. [00:26:04] Speaker 00: They have to use what's on the NIC and the control node. [00:26:07] Speaker 02: So why wouldn't it be a [00:26:11] Speaker 02: reasonable question for a jury, whether it's reasonable to say that when a processor uses the emulation that we apparently all agree is done at the NICS, that that is the processor emulating as well. [00:26:34] Speaker 02: Why is that not just a question of a reasonable application of that construction? [00:26:40] Speaker 00: So I guess two responses. [00:26:41] Speaker 00: One is, for the same reason, it's not reasonable for me to say, I am part of the phone network when I pick up, when I dial a phone number on my phone. [00:26:49] Speaker 00: My phone is part of the network, and the phone company has created the network. [00:26:53] Speaker 00: So that's just, that's not a reasonable interpretation of emulation. [00:26:58] Speaker 00: And the second point, Your Honor, is that these claims have these really specific limitations because eGenera wants all this networking functionality [00:27:07] Speaker 00: and the ethernet functionality to be on the processors, because they're trying to create processor networks. [00:27:12] Speaker 00: Cisco's system is just fundamentally different from that, in that it's creating the server networks, and that's why we put all of this functionality on the NICs. [00:27:21] Speaker 00: We proved that backwards and forwards at trial, when we had a lot more space and evidence. [00:27:28] Speaker 00: And that is reflected in all of Egenera's specification. [00:27:35] Speaker 00: If we look at the specification at appendix 103, it talks about how the processor logic, so this is column four, lines 56 to 63. [00:27:46] Speaker 00: That portion of the specification specifically talks about how the processors include logic to emulate the semantics of an ethernet driver. [00:27:59] Speaker 00: So this whole specification, which describes [00:28:04] Speaker 00: Eugenia has admitted describes their blade frame product. [00:28:07] Speaker 00: This Ethernet emulation is in here because they want to use the GigaNet protocol for their internal communications. [00:28:15] Speaker 00: So they are requiring that these drivers, the Ethernet drivers, the Ethernet programming actually be on the processors, not just that the processors use Ethernet functionality. [00:28:26] Speaker 04: Is it clear on this record that the Ethernet driver sits on the VNIC or sits [00:28:33] Speaker 04: Or is it on CPU? [00:28:35] Speaker 00: It's on the VNIC. [00:28:38] Speaker 00: I think that's an appendix 4127 is the expert evidence on that. [00:28:43] Speaker 04: When the CPU sends a message to the VNIC, through the VNIC output or to the VNIC, it doesn't have a... And it's using Ethernet terminology, right? [00:28:55] Speaker 04: The right headers, basically. [00:28:58] Speaker 00: There are Ethernet tags. [00:29:00] Speaker 04: Do you have a driver to do that? [00:29:04] Speaker 00: I'm not sure, Your Honor. [00:29:06] Speaker 02: That's one question on the new trial motion. [00:29:10] Speaker 02: You say that a lot of things that they complain about having gone awry at trial, I guess, reference to the empty chair witness litigation finance discussion in a closing argument, maybe things that were said about them as an NPE that were sort of arguably negative, that they only went to damages [00:29:33] Speaker 02: And therefore, the jury didn't reach damages. [00:29:37] Speaker 02: So it's not relevant. [00:29:39] Speaker 02: But isn't the concern that all those things might have created such a cloud that it distracted the jury and was prejudicial to them by distracting the jury from the actual evidence? [00:29:52] Speaker 00: So just to talk through sort of the layers of review that are happening here. [00:29:57] Speaker 00: So the district court has reviewed his motions in Lemonade and said, [00:30:02] Speaker 00: there wasn't a clear violation of the motion in Lumine. [00:30:04] Speaker 00: So first we'd have to say, there actually was a violation, and you're interpreting your own orders wrong. [00:30:10] Speaker 00: Then eGeneric didn't object to that evidence, either, or that argument. [00:30:14] Speaker 00: So then we have to say, it doesn't matter that they didn't object, or even raise this after the closing arguments, that there's still some kind of plain error. [00:30:24] Speaker 00: And then we can get to whether there was prejudice from these statements. [00:30:29] Speaker 00: Cisco was really clear in its closing argument that it was linking those statements to damages, and this idea that there was a cloud over Egenera, Egenera put forth its BladeFrame product as a patent-practicing product, and it tried to say Cisco copied the BladeFrame, and then when Cisco introduced its UCS system, that's what tanked Egenera's sales and why Egenera no longer practices this patent. [00:30:55] Speaker 00: Cisco thoroughly disproved that at trial. [00:30:58] Speaker 00: We proved there was no copying, that the BladeFrame product was failing in the market well before UCS was introduced, and that Egenera's decision to stop selling the BladeFrame had nothing to do with Cisco's UCS system. [00:31:11] Speaker 00: So it was really important for us to respond to Egenera's narrative by explaining why BladeFrame came off the market and why that had nothing to do with Cisco, but had everything to do with Egenera's own business decisions. [00:31:29] Speaker 01: Thank you very much. [00:31:30] Speaker 01: Thank you. [00:31:38] Speaker 03: I don't have anything further to add. [00:31:39] Speaker 03: If you'd like to ask me a question, I'm happy to answer it. [00:31:41] Speaker 03: But otherwise, thank you. [00:31:43] Speaker 03: Thank you. [00:31:44] Speaker 01: We thank both sides.